Known switch systems with a telecommunications network typically have many service circuits each one of which is connected to many types of service circuit resources. Service circuits perform the basic functions that are required for every call. Service circuit resources are specialized circuits (e.g., speech detection circuitry and video conference bridges) that perform functions not required for every call.
For example, U.S. Pat. No. 5,440,563 by Isidoro et al., issued on Aug. 8, 1995 and assigned on its face to AT&T Corp., is an example of a known switch system. The switch allocates a pool of service circuits among many switches in a telecommunications network. A regional control node external from the switches controls the connection of the service circuits to the switches. The individual service circuit resources are collectively allocated with the service circuit itself.
These known switch systems, however, suffer several shortcomings. First, requiring more service circuit resources than is required can be quite expensive. Although a subscriber may choose to have a particular resource available for calls, not all calls require that resource to be connected for the entire duration of the call. Even when the call processing and/or service logic of the switch turn off a resource, the resource remains assigned to the call and cannot be reassigned to any different calls.
For example, although a subscriber may choose to have speech detection capabilities available to callers, when a caller opts to interact with the switch through dual tone multi-frequency (DTMF) signal rather than through speech, the speech detection resource is no longer required to be connected to the call. In such a case, the resource is tied up unused until that call is terminated. Even if the speech detection resource is turned off by call processing and/or service logic, the speech detection resource is unavailable to be used by other calls. This reduces the efficiency of utilization of these resources and thereby increases the costs associated with the switch and, consequently, with the call.
Second, when a particular resource fails, known switches cannot distinguish this failed resource from functioning resources connected to the same service circuit. Consequently, more resources are removed from service for repair and/or diagnosis than is necessary.